Social capital across different residential communities in china

At the individual level, it can be manifested in one‟s social network, trustworthiness, social status, and so on; at the group level, it is embodied by social norms, trust, etc., which f

SOCIAL CAPITAL ACROSS DIFFERENT

RESIDENTIAL COMMUNITIES IN CHINA

YUAN XU

NATIONAL UNIVERSITY OF SINGAPORE

2011

SOCIAL CAPITAL ACROSS DIFFERENT

RESIDENTIAL COMMUNITIES IN CHINA

YUAN XU

(B.M.), Central University of Finance and Economics

A THESIS SUBMITTED

FOR THE DEGREE OF MASTER OF SCIENCE

(REAL ESTATE AND URBAN ECONOMICS)

DEPARTMENT OF REAL ESTATE

SCHOOL OF DESIGN AND ENVIRONMENT

NATIONAL UNIVERSITY OF SINGAPORE

2011

Acknowledgement

This thesis could not have been finished without the help of a number of people Particularly, I would like to take this opportunity to express my sincere thanks to the following:

Firstly, I would like to express my sincere thanks to my supervisor, Professor

Fu Yuming, Vice Dean of the School of Design and Environment, National University of Singapore Professor Fu has generously spent time on guiding

me, from choosing modules, to finding a research topic, to conducting empirical analysis, to writing my thesis, to improving my presentation skills His encouragement and patient guidance have not only helped me finish my master study, but also changed my attitude towards work and life in the future Due to the supervision of Professor Fu, I believe I have improved a great deal as a researcher and as an individual

Secondly, I would like to thank Professor Tu Yong and Professor Ong Seow

Eng The modules of Research Methodology in Real Estate and Real Estate

Economics Seminar taught by Professor Tu, and the module of Research Topics in Real Estate jointly given by Professors Tu and Ong have equipped

me with a solid foundation of doing research

Thirdly, I want to thank Ms Zainab Binte Abdul Ghani and Ms Nor‟Aini Binte Ali for patiently offering me with a lot of help in relation to the administrative matters I encountered in my master study at the National University of Singapore

Fourthly, I want to thank the two anonymous examiners for providing valuable suggestions of improving my thesis I have benefited a lot from the constructive advice

I also wanted to say a “thank you” to my seniors and friends; especially Li Pei, Liang Lanfeng, Xu Yiqin, Wang Yourong, Guo Yan, Zhang Liang, Li Qing, and Zhao Yibo among others, who have offered valuable direct and indirect suggestions for my study and research, and have offered me a lot of help in my life in Singapore

Last but not least, I want to thank my family members, especially my parents and my little sister I thank my parents for supporting my education for so many years and my choice of studying at the National University of Singapore, and I thank my sister for being a very good daughter at home and for providing company to my parents while I am away

Thank you all!

Table of Contents

Acknowledgement i

Table of Contents iii

Summary v

List of Tables viii

List of Appendices x

1 Introduction 1

1.1 Research Motivation 1

1.2 Thesis Organization 6

2 Literature Review 9

2.1 Review on Social Capital Literature 9

2.2 Review of Social Interactions Literature 13

2.3 Chinese Literature Review 16

3 Institutional Background 19

3.1 The Chinese Culture 19

3.2 Formation of Different Residential Community Types 20

3.3 The Hukou System in China 24

4 Methodology 26

5 Data 29

5.1 Description of the Data Set and Variables 29

5.2 Descriptive Statistics of Key Variables 35

6 Empirical Result 40

6.1 Separation of Private Effects and Social Influence 40

6.2 Determinants of Social Influence 45

7 Conclusion 59

7.1 Main Findings and Policy Implications 59

7.2 Limitations of This Study and Suggestions for Further Research 61

References 64

Appendices 73

Summary

Social Capital is an intangible asset residing in social relations, which could help improve the welfare of an individual and the group the individual belongs to At the individual level, it can be manifested in one‟s social network, trustworthiness, social status, and so on; at the group level, it is embodied by social norms, trust, etc., which facilitate cooperation, social cohesion and civic engagement and therefore help improve social and economic outcomes

In academic research, social capital has been viewed as partly a private good,

as it could be built up by investment; in the meantime, it may also be viewed

as a public good, as it produces external benefits through social influences Moreover, social capital has been widely recognized as an important factor facilitating collective actions and social economic development as well as promoting individual success and well-being (e.g Glaeser, Laibson, and Sacerdote,2002; Durlauf and Fafchamps, 2005; Helliwell and Wang, 2010)

A residential community or neighborhood is a place where individuals may spend most of their spare time, and therefore, the social capital within a residential community is likely to be an important concern of improving residents‟ wellbeing Although the location-specific nature of social capital

in a residential context has been closely examined in developed economies

like the US, little research has been conducted in China, the largest emerging economy of oriental culture

Using the China General Social Survey 2005 (CGSS 2005) and other sources

of data, this study attempts to examine how individual characteristics and location attributes influence individual social capital, measured by four indicators the extent to which individuals are acquainted with (ACQUAINT) and trust their neighbors (TRUST), and to which they are interested and involved in community affairs (INVOLVE), and the frequency

at which they exchange help with their neighbors (HELP) Among these four social capital indicators, ACQUAINT and TRUST represent individual‟s perception of their social capital within the residential community, while INVOLVE and HELP could be viewed as their action of investing in local social capital

After reviewing relevant theories, the empirical part is composed of two stages: (i) the decomposition of the variation in social capital indicators into individual effects and social influences (the community fixed effects); (ii) determinants of the social influences Empirical results for the first stage show that individuals‟ perceptions of their neighbors‟ trustworthiness are the most difficult to explain; getting older, becoming a Communist party member and a homeowner seem to improve all of the four social capital

indicators of ACQUAINT, TRUST, HELP and INVOLVE Variance decomposition after the first stage analysis shows that the variation is overwhelmingly due to differential social influences, as opposed to differential individual attributes Analysis of the community fixed effects of ACQUAINT, TRUST, HELP and INVOLVE reveals more positive social influences in rural communities, including rural villages and towns and

“urban villages” At the city level, local social capital is higher for communities located in cities with a lower average annual disposable income and better climate, measured by higher temperature index and lower humidity

Conclusions obtained from this study highlight the need for policy attention

to manage the potential decline in social capital, with rising affluence and income inequality in Chinese cities, at a time when social capital is much in need for building new urban communities Measures should be taken to encourage social interactions in residential neighborhoods that promote social capital Moreover, the importance of urban attributes in influencing social capital shows the need for more research in the future to improve our understanding of the role of the urban environment in social capital formation

List of Tables

TABLE i Description of Individual Variables from CGSS2005 (9406

Observations) 31

TABLE ii Correlation Matrix of Individual Variables (9406 Observations) 32

TABLE iii Description of Community Variables from CGSS2005(348 Observations) 33

TABLE iv Correlation Matrix of Community Variables from CGSS2005(348 Observations) 33

TABLE v Description of Urban Variables [number in the brackets indicates source] (49 to 54 observations) 34

TABLE vi Descriptive Statistics of Key Variables 36

TABLE vii Separation of Private Effects and Fixed Effects 44

TABLE viii Variance Decomposition 45

TABLE ix Mean and Standard Deviation of Social Influences and Community Average Individual Effect 47

TABLE x Correlation Matrix of Social Influences 48

TABLE xi Variation of Local Social Capital across Different Residential Community Types 49

TABLE xii OLS Estimates of Cross-Community Determinants of Social Influences on Social Capital 50TABLE xiii OLS Estimates of Cross-community Determinants of

Social Influences on Social Capital (Continued) 51TABLE xiv OLS Estimates of Cross-Community Determinants of Social Influences on Social Capital 56TABLE xv OLS Estimates of Cross-Community Determinants of Social Influences on Social Capital (Continued) 57

1 Introduction

1.1 Research Motivation

Social Capital is an intangible asset residing in social relations, which could help improve the welfare of an individual and the group the individual belongs to At the individual level, it can be manifested in one‟s social network, trustworthiness, social status, and so on; at the group level, it is embodied by social norms, trust, etc., which facilitate cooperation, social cohesion and civic engagement and therefore help improve social and economic outcomes

At the individual level, it can be manifested in one‟s social network, trustworthiness, social status, and so on; at the group level, it is embodied by social norms, trust, etc., which facilitate cooperation, social cohesion and civic engagement In academic research, social capital has been viewed as partly a private good, as it could be built up by investment(Glaeser, Laibson and Sacerdote, 2002); in the meantime, it may also be viewed as a public good, as it produces external benefits manifested through social influences(Durlauf and Fafchamps, 2005) Moreover, it has been widely recognized as an important factor facilitating collective actions and social economic development as well as promoting individual success and

well-being (Glaeser, Laibson, and Sacerdote, 2002; Durlauf and Fafchamps, 2005; Helliwell and Wang, 2010)

A residential community or neighborhood is a place where individuals may spend most of their spare time, and therefore, the social capital within a residential community is likely to be an important concern of improving residents‟ wellbeing Researchers have studied social capital in a residential community based in a western context The location-specific nature of social capital has been examined in terms of its interaction with residential mobility (Glaeser and Redlick, 2008; Kan, 2007), its influence on citizenship and community governance (DiPasquale and Glaeser, 1999; Olken, 2009), and its rise and decline with residential community changes and urban sprawl (Hilber, 2010; Brueckner and Largey, 2008) Although the location-specific nature of social capital in a residential context has been closely examined in developed economies, little research has been conducted in China, the largest emerging economy of oriental culture

Compared with countries with a western culture, the Chinese society has its unique characteristics, because the Chinese strongly emphasize on

“guanxi”(it means social relationship or social network in Chinese) and the transitional nature of the Chinese society In China, not only did ordinary Chinese pay attention to cultivating their “guanxi” to positively influence

their job hunting outcomes either by providing more information or directly affecting the results(Bian and Zhang, 2001; Chen and Wang, 2004), companies also refer to “guanxi” to assist their market expansion and competitive position(Bian and Qiu, 2000) What‟s more, Chinese residential communities have undergone great changes in the past several decades Hundreds of millions of Chinese people moved from rural areas to cities and few urban neighborhoods have been left untouched by urban redevelopment

On the other hand, the housing policies carried out by the Chinese government have further influenced the formation of current urban communities by implementing the commercial housing policy Over three decades ago, nearly all urban homes were state owned and managed; today, the vast majority of urban homes is privately owned and depends on private efforts and initiatives for community management and well-being Urbanization and the implementation of commercial housing policies entail social structural change, which affects social capital At the same time, social capital would have an important role in shaping the healthy social development in the emerging and transitioning residential neighborhoods in Chinese cities The strengthening or weakening of social capital in residential communities should be important policy concerns

The fact that there is relatively limited research conducted based on the unique background of China, calls for further study of analyzing the variation in

social capital at the individual level and across residential communities, and the present study is an attempt to fill this research gap and provide more empirical evidence of social capital in a transitional economy like China This study employs the 2005 China General Social Survey (CGSS2005) as its primary data source CGSS2005 samples clusters of residents by residential communities in over 100 cities, whose political boundaries typically cover both rural and urban districts In this study, a residential community is delineated by an urban residential committee (called Juweihui in Chinese, which includes about a thousand households) boundary or by a rural village committee (Cunweihui) boundary In addition to CGSS2005, this study also employs information of urban attributes from the 2007 Urban Household Survey, Chinese Statistical Yearbooks, and so on

Although consensus has been reached that social capital is generally indicated

by individuals‟ social networks and trust; and socially it is manifested in enhanced economic exchange and public goods, yet there is no precise way of how to quantitatively measure social capital Therefore, four social capital indicators are chosen for this study based on the definition: (i) how widely they trust their neighbors (TRUST), (ii) how acquainted they are with their neighbors (ACQUAINT), (iii) how often they exchange help with their neighbors (HELP), and (iv) how interested and involved they are in the work and decisions of their residential or village committees (INVOLVE) These

indicators are reported in CGSS2005 based on individuals‟ self-assessments Among these four measures, TRUST and ACQUAINT can be considered as indicators of the individual‟s trust and social connection, while HELP and INVOLVE reflect the individual‟s willingness to contribute to the economic exchange and public goods in the community

One focus of this study is placed on examining how individual characteristics influence the individual social capital indicators, namely TRUST, ACQUAINT, HELP and INVOLVE Empirical results suggest that possessing hukou and becoming a homeowner increase individuals‟ investment in social capital TRUST is the most difficult to explain among the four social capital indicators Variance decomposition of the first stage estimation suggests that, whereas the private motivations have noticeable influences on individuals‟ social capital indicators, the social influences appear to have far greater importance

The second focus of the present study is placed on the social influences at the residential community level This study identifies the social influences as community fixed effects on individuals‟ assessments of TRUST, ACQUAINT, HELP and INVOLVE To examine the contributing factors of the social influences at the residential community level, three empirical estimation methods are adopted: the first approach uses all of the sample in the first stage

regression to see the relationship between residential community types and the local social capital measured by SI_TRUST, SI_ACQUAINT, SI_HELP and SI_INVOLVE respectively; the second and third approaches only consider the residential communities in the urban areas of cities with populations greater than 500,000 to see the effect of residential community types, community attributes and city attributes on the four local social capital indicators Empirical results from the second stage estimation show that social capital is higher in rural villages and towns and “urban villages” but lower in commodity-housing communities In addition, local social capital tends to be lower in more affluent cities and in cities with greater education and income disparity within residential communities

To the extent that social capital contributes to community well-being, the findings of the present study highlight the need for policy attention to manage the potential decline in social capital, with rising affluence and widening inequality in income and education in Chinese cities, at a time when social capital is much in need for building new urban communities

1.2 Thesis Organization

The thesis is organized as follows:

 Chapter 1 contains a brief overview of the research background, research problems, research objectives and empirical results It also introduces the organization of this study

 Chapter 2 reviews international and Chinese literature to support the theoretical part and the empirical methods to be employed International literature is composed of social capital literature and social interactions literature The social capital literature provides the theoretical founding for this study, while the social interaction literature provides guidance for empirical analysis The Chinese literature highlights the knowledge gap and demonstrates the need for, and significance of, this study

 Chapter 3 introduces the Chinese social background, including the formation of the different residential community types in China, the Chinese hukou system, and the characteristics of Chinese culture

 Chapter 4 presents the empirical methodology for the separation of private effects and social influences in the first stage analysis and for the determinants of the social influence in the second stage analysis

 Chapter 5 describes the variables to be used in the empirical analysis, and then provides descriptive statistics of the variables to be used

 Chapter 6 is divided into two stages: the decomposition of private effects and community-specific predicted values due to social influence and the determinants of the social influence In the first stage, the entire sample

is used to estimate the influence of individual characteristics and location fixed effects The second stage estimation is designed to explore the differences of local social capital at the community level and city level, and different samples are used

 Chapter 7 concludes with a discussion on the main results obtained, policy implications, limitations of this study and suggestions for future research

2 Literature Review

2.1 Review on Social Capital Literature

After Loury (1977) introduced the research topic of social capital into modern social science research and Coleman‟s (1988) seminal study placed it at the forefront of research in sociology, the term social capital has spread throughout the social sciences

In sociology, the academic interest in social interactions starts from the 1990s with the work of Jencks & Mayer (1990, Mayer & Jencks 1989) They argue that if growing up in a poor neighborhood matters, intervening processes such

as collective socialization, peer-group influence, and institutional capacity could be part of the reason After that, many topics related to neighborhood effects have been studied, e.g neighborhood disadvantage, child and adolescent outcomes, health indicators, etc According to a review by Sampson, et al (2002), several consensuses have been reached in social capital studies at the residential community level: (1) considerable social inequality among neighborhoods in terms of socioeconomic and racial segregation exists; (2) a large number of social problems tend to come out together at the neighborhood level; (3) two sets of clusters are related—neighborhood predictors and length of tenure; (4) empirical results

have not varied much with the operational unit of analysis; (5) the ecological concentration of poverty appears to have increased significantly during recent decades

In economics, Helliwell and Putnam (2000) examined relationship of social capital and the per capita output growth in regions in Italy; they found that higher social capital was associated with higher growth Charles and Kline (2006) found that ethnic heterogeneity reduces social capital formation for some pairings, notably whites and blacks and whites and Hispanics Especially, Researchers specializing in real estate and housing economics have also studied various topics of neighborhood effects at the community level, covering topics related to housing maintenance behavior, individual employment, children‟s education, housing demand, property valuation, mortgage default, etc Davis and Whinston (1961), Rothenberg (1967), Stahl (1974) and Schall (1976) studied housing maintenance behavior, in which Stahl (1974) is generally perceived to be the first formal model of residential neighborhood interactions in the context of housing decisions that invoke a symmetric Nash equilibrium setting Schelling (1971) analyzes the residential patterns that emerge when individuals choose not to live in neighborhoods where the percentage of residents of their own race is below a certain threshold Ioannides (2003) studies neighborhood effects with regard to property valuations, and he concludes that the endogenous social interactions

range from 0.58 to 0.77 Ioannides and Zabel (2003) study neighborhood effects and housing demand, and Ioannides and Zabel (2008) further study this topic by controlling for the individual sorting, finding that individuals prefer to live near others like themselves Oreopoulos (2003) studies the effect of living

in a poor neighborhood on the labor market outcomes and finds that neighborhood quality plays little role in determining youth‟s eventual earnings, unemployment likelihood and welfare participation Goux and Maurin (2007) study the influence of close neighbors on children‟s early performance at school and conclude that adolescents‟ outcomes at the end of junior high school are strongly influenced by the performance of other adolescents in the neighborhood Dujardin and Goffette-Nagot (2010) study neighborhood effects on unemployment Using the data from the Social Capital Community Benchmark Survey, Hilber (2010) further extends the result of DiPasquale and Glaeser (2000) and finds that the positive linkage between homeownership and housing supply is confined to built-up communities, because homeownership prevents newcomers from coming into the community and upsetting a mutually beneficial equilibrium involving reciprocal cooperation between neighbors David et al (2010) also conclude that more social capital

is associated with higher mobility within a country and at the individual level using the European data Also, in the study field of real estate finance,

Ambrose et al (2009) find that living in a subprime neighborhood increases the individual households‟ possibility of default

Although there has been a huge amount of literature in this area, the definition

of social capital still suffers from the problem of vagueness (Durlauf, 2000; Manski, 2000) For example, Putnam et al (1993) provide a definition of social capital, “…social capital…refers to features of social organization, such as trust, norms, and networks that can improve the efficiency of society…” In later research, Putnam (2000) defines social capital as “ connections among individuals - social networks and the norms of reciprocity and trustworthiness that arise from them.”

There are still other definitions provided by researchers, however, three common features have been emphasized: (1) positive externalities for members of a group; (2) these externalities are achieved through shared trust, norms, and values and their consequent effects on expectations and behavior; (3) shared trust, norms, and values arise from informal forms of organizations based on social networks and associations Consequently, research on social capital should try to emphasize all three, or at least two, of the points above (Durlauf, 2005) A great number of studies related to social capital use a certain behavior as a proxy for social capital For instance, DiPasquale and Glaeser (1999) use a range of citizenship variables (voting in local elections,

helping solve local problems, knowing school principals, etc.) as proxies of social capital Helliwell and Putnam (2000) used civic community (index of associations, newspaper readership, and political behavior), institutional performance, and citizen satisfaction with government as social capital measures; Charles and Kline (2006) used carpooling as the proxy for social capital In this study, the four social capital indicators—TRUST, ACQUAINT, HELP and INVOLVE are chosen due to the purpose of trying to capture all these three characteristics of social help Specially, among these four social capital indicators, HELP and INVOLVE has captured the positive externalities for a group and the consequent behavior of a group member; TRUST has captured the shared trust of the group; ACQUAINT has picked up the social network and the positive externalities of a group

2.2 Review of Social Interactions Literature

There are three branches of literature in neoclassic economic theory providing theoretical foundation for social interactions: game theory in microeconomics, social interaction and labor economics, and endogenous growth theory in macroeconomics In decision and game theory, Conlisk (1980) shows that imitators may have a high long-run fitness as the optimizers, so if the decision-making process is costly, it may be optimal for individuals to imitate behaviors of other persons who are better informed Akerlof (1980) examines the adherence to social customs, and finds that if an

individual disobeys the social customs the individual will be punished, while the social customs persist without erosion Bernheim (1994) develops a model of social interactions in which individuals care about both social status and intrinsic utility When social status is sufficiently more important than intrinsic utility, many individuals conform to the social norms If an individual deviates from the social norms, he or she will be punished and the penalty is produced endogenously The above game theory studies all point out that conformity will increase individual utility while deviation will reduce utility from different angles, while studies in social interactions and labor market studies make the consequence of social interactions evident In

a study related to social interactions and the labor market, Rees et al (2003) and Mas and Moretti (2009) study the influence of peers on individuals‟ performance at work In the macroeconomic endogenous growth model, two critical drivers for the economic growth are knowledge spillover and technological development, both of which are likely to be associated with social interactions For example, Berliant et al (2006) studies knowledge exchange and economic development

Glaeser and Scheinkman (2003) provide a general model and a broad empirical overview about the research on social and nonmarket interactions, and they introduce a general theoretical model for social interactions The

group , the preference shock , which is determined by the agent‟s characteristics, like income, education and so on, and p the cost of taking the action

Glaeser and Scheinkman (1999) have shown a specific example of how individual choice, reference group choice, individual characteristics and cost

of taking action influence the individual utility,

, (1)

measures the taste for conformity There are three channels for conformity: (1) the endogenous effects, which is that the propensity of an individual to behave in some way varies with the group behavior; (2) contextual effects, which is that the propensity of an individual

to behave in some way varies with the characteristics of the group; (3) correlated effects, which means that individuals behave similarly in the same group because of the (unobserved) similar personal characteristics they have and the similar institutional characteristics they are confronted with (Manski, 1993)

Equation (1) indicates that the closer the individual choice is to the group average choice, the higher utility the agent will obtain This utility function gives the following equilibrium condition,

(2)

Equation (2) is a relatively simple liner model to describe the determinants of social capital and have laid a foundation for the empirical part of this study

2.3 Chinese Literature Review

Due to the important role played by guanxi in the Chinese society, variables related to social networks or guanxi are usually used as the indicator for social capital in most Chinese social capital literatures, and various relevant topics are examined, including factors contributing to individual social capital, and the effects of social capital on different social and economic outcomes both at the individual level and at the corporate and institutional level, and so on

At the individual level, Bian (2004) surveyed the sources and consequences

of social capital online, and finds that occupation and income affect individual social capital, and that individual social capital could benefit individuals in improving trust, reinforcing reciprocity, building reputation and increasing the quality and quantity of relevant information In addition, a number of studies have been conducted to examine the influence of social capital on employment, targeting a group of people with certain characteristics, such as laid-off workers (Zhao and Feng, 2000; Gui, et al 2002), farmers (Chen and Wang, 2004), and recent college and university

graduates (Zheng, 2004; Chen, Zheng and Wu, 2005) Empirical researches

on employment suggests that social networks described by relatives and friends could have direct impact on job hunting outcomes; while social networks described by acquaintances helps provide job information and therefore have indirect impact on job hunting results (Bian and Zhang, 2001; Chen and Wang, 2004)

At the institutional level, guanxi is intensively studied, covering the relationship between guanxi and institution and economy growth, the role played by guanxi on the advancement of governing systems, the effect of guanxi on corporate performance, and so on In research related to economic institution development, researchers argue that guanxi is a necessary step in the process of a country building a just and efficient background institution; once the legal and economic systems are fully developed, the need for guanxi will decrease or disappear (Xin and Pearce, 1996; Peng and Luo, 2000; Luo, 2001; Park and Luo, 2001) In addition, guanxi could be a double-edged sword in its relationship with the governing system For example, guanxi could ensure that companies will donate money for environmental clean-up and waste disposal (Warren, et.al, 2004), but on the other hand, guanxi prevents the implementation of China‟s environmental policy because it is in the way of enforcement (Swanson et al, 2001)

At the corporate level, research has been conducted to examine the influence

of social capital on company performances Bian and Qiu (2000) conducted a survey of over 100 companies in Guangzhou, and concluded that social capital possessed by companies helps to improve company performance Park and Luo (2001) find that although guanxi has limited influence on sales growth and profit growth of the company, it does lead to better firm performance and could also assist the market expansion and competitive positioning of firms In addition to the positive impacts of social network when seeking a job, researchers also noticed the negative influences of social capital on the job market, like excluding others who do not have strong social networks and causing consequent social inequality

Although a number of studies have been conducted in China, few of the conclusions are based on empirical estimation results, and even fewer try to study social capital at the residential community level In addition, the social capital indicators chosen are limited These call for further study on the topic

of local social capital at the residential community level

3 Institutional Background

3.1 The Chinese Culture

The Chinese have long believed in the role played by social influence and social capital One example of social influence is an old Chinese saying that

“Keep good men accompany and you shall be of the number”, and it means that due to social influences, individuals could further improve

Another characteristic of Chinese Society is the strong emphasis on “guanxi” which means social network or social relationship and which is part of the

“social capital” discussed in this paper Individual Chinese refer to “guanxi” for their job hunting and Chinese firms refer to “guanxi” to seek benefits for the firm as well Therefore, Chinese people might be more willing to offer help to their neighbors to enhance their reciprocal social network; or the wealthier people and people with higher education may be more willing to offer help compared with those with lower income or lower education, because the wealthier people and people with higher education may have benefited more frequently from their “guanxi”

The strong emphasis on social influences and “guanxi” in the Chinese society not only hints that the social capital might be more important in China, but

also shows that the formation of social capital might be different compared with other countries or cultures

3.2 Formation of Different Residential Community Types

In mainland China, residential communities have undergone great changes in the past several decades due to both the urbanization process and the housing policy change Till now, five main residential community types existed: (1) old downtown communities; (2) working-units communalities; (3) commodity-housing communities; (4) towns and urban villages; (5) rural communities Among these five residential community types, old downtown communities have a long history and diverse residents, working-unit communities have residents with a relatively homogeneous background, commodity-housing communities are relatively new and have residents with higher mobility, towns and urban villages may result from informal housing development, and rural villages also have a long history and have residents, most of whom make their living from farming The different characteristics of these five different community types in China might bring about different ways of social capital formation The sample of CGSS2005 covers all of these community types

The urbanization process of China started after the foundation of the People‟s Republic of China and grew rapidly after the proposal of the “Reform and

Opening Up” policy at the end of 1978 The urban population grew steadily at

an annual rate of around 3-4% from 1950 to 1965 The urban population experienced a great jump from 1958 to 1961 during the “Great Leap

years from 1965 to 1975, urban population growth rate dropped substantially as a result of rustication of youth and other social

launched at the end of 1978, urban population growth began to accelerate The inflow of foreign direct investment created massive employment opportunities, which fostered urban population growth From 1978 to 2002, the urbanization rate in China increased from 17.92% to 39.1% The annual growth rate was

26% in 1990 and 36% in 2001 At the end of 2009, China's total population

1

The Great Leap Forward of the People‟s Republic of China was an economic and social plan used from 1958 to 1961, which ostensibly aimed to use China‟s large population to rapidly transform the country from an agrarian economy into a modern communist society through the process of agriculturalization, industrialization, and collectivization Zedong Mao(the chairman of the communist party of China that time) led the campaign based on the Theory of Productive Forces, and intensified it after being informed of the impending disaster from grain shortages.

4

From the Statistics of the Ministry of Construction

http://english.peopledaily.com.cn/200405/19/eng20040519_143708.html

was 1.334 billion, with 712 million (53.4%) and 622 million (46.6%) residing

in rural areas and urban areas respectively (not including Hong Kong, Macau,

previously lived in rural areas now live in the city, either because they moved

to the city or the place they previously lived was transformed from a rural area

to an urban area The communities they previously lived in have also changed

In traditional Chinese villages, a large family usually lived together, and they and their neighbors were likely to have lived in the same village for generations However, after the urbanization of the village, many outside residents began to move in and the villagers began to move out, so that the socioeconomic composition of the neighborhood was changed

The urbanization process mainly concerns the change in the rural communities, while the housing policy change mainly concerns residential communities in the urban areas The public housing system, which took effect after the foundation of the People‟s Republic of China, has provided Chinese urban dwellers with low-cost accommodation However, due to the rapid urban population growth caused by the urbanization process, and the lack of systematic planning, great problems have been caused by the public housing system, such as shortage of housing, prevalence of dilapidated housing,

housing inequality, heavy burden on the state, construction of low-quality

5

From the report by the Chinese Academy of Social Science

http://www.cpirc.org.cn/news/rkxw_gn_detail.asp?id=10684

housing, inappropriate maintenance, irrational pattern of ownership, and so on The housing reform that surfaced in 1978 under the leadership of Xiaoping Deng is designed to initiate a rational mechanism for housing investment, construction, management and distribution by changing the public housing system to a private - ownership housing system, and changing the system where housing is provided by the state and the working units to a system where housing costs are shared by the state, working units and individual households (the state and working units provide the housing benefits via the

by the state and the working units, urban workers in the same working units lived together, so they knew each other well, and were willing to offer help when other neighbors were in trouble Now, due to the construction of commercial housing and the fact that working units stopped providing

10

The commercial mechanism of the housing system began to be constructed, and various policies took effect to make the system better and to deal with problems arising from the process of developing the market dominated housing system See Shaw (1997).

apartments, urban workers could not live together and they began to live with other people they did not know

3.3 The Hukou System in China

The hukou system, which is the system of residency permits, dates back to the 1950s In China, every household has a Household Registration Record, which includes various information of family members, such as name, birth, death, identification number, residence location, moves, and so on Due to the differences in the living conditions, education and healthcare opportunities between the rural and urban areas, the hukou system is used to limit mass migration from rural areas to urban areas in order to ensure the structural stability of society It may also be used to monitor dubious criminals to ensure the safety and stability of society

Before the implementation of the Opening and Reform Policy in the late1970s

in China, the police would periodically round up those farmers who were in the city and send them back to their permanent residency locations After the late 1970s, due to the fact that the development of the coastal areas cause a need for workers, the farmers in the rural areas and workers in small and medium cities are allowed to work in these coastal cities, but they still need a working permit to work in the city Although it seems that the hukou system is not that strict compared with the past, since residents from the middle and

western regions could work and live in these coastal cities if they have enough money to support themselves, they still could not enjoy the sound healthcare system in the city, their children could not have the same education opportunities as children of urban residents, and also in some cities like Beijing and Shanghai, they could not buy an apartment if they have not paid more than five years‟ income tax Therefore, the hukou status has been an important component of Chinese life To some extent, it represents the rights

of enjoying a better life and more opportunities in the place where you have hukou

Therefore, to some extent, whether one has hukou in the community or not could represent how long one is going to stay in the community and how much one is going to invest in the local social capital of the community, because if one does not have hukou in the community, he or she does not intend to stay

in the community for a long time or he or she will be forced to leave by his or her inability of enjoying the local welfare there

4 Methodology

This study adopts a linear model of social interactions (Durlauf and Ioannides, 2010) to describe the determination of the choice or propensity

study—of individual i in social group j:

unobserved individual heterogeneities For the purpose of the present study, this paper identifies the social group with the residential community where

individual i lives

Further, this study assumes μ ij to have a mean μ j, reflecting the possible

correlation of unobserved individual heterogeneities within the community

due to endogenous sorting, and a standard error σ j ; thus μ ij =μ j +ε ij , where ε ij

The α coefficients determine the private effect due to individual attributes, whereas the β coefficient captures the endogenous effect

Substituting the expected value of y ij at the community level for m j in

Equation (3), the following equation is obtained:

= (4)

where the terms within the square brackets represent different sources of the

location effect, and the correlation effect, respectively

The empirical analysis proceeds in two steps In the first step, Equation (4) is estimated to identify the terms in the square brackets as the community fixed

which is not necessarily linear and is potentially subject to censoring due to the survey instrument, Equation (4) is estimated using an ordered-probit model

at discrete values of 0, 1, …, H, with H being the highest reported level for

the indicator The ordered-probit model estimates the equation for the latent

relationship:

=0, if

(5)

y ij is linear and not censored

In the second step, the estimated community fixed effect γ j is regressed on

The ordered-probit model assumes ε ij to follow a normal distribution with mean zero

and variance σ2; hence the probability of observing Y ij = h, 0 ≤ h ≤ H, is given by

( )